A Monte Carlo and experimental investigation of the dosimetric behavior of low‐ and medium‐perturbation diodes used for entrance in vivo dosimetry in megavoltage photon beams

Full buildup diodes can cause significant dose perturbation if they are used on most or all of radiotherapy fractions. Given the importance of frequent in vivo measurements in complex treatments, using thin buildup (low-perturbation) diodes instead is gathering interest. However, such diodes are strictly unsuitable for high-energy photons; therefore, their use requires evaluation and careful measurement of correction factors (CFs). There is little published data on such factors for low-perturbation diodes, and none on diode characterization for 9 MV X-rays. We report on MCNP4c Monte Carlo models of low-perturbation (EDD5) and medium-perturbation (EDP10) diodes, and a comparison of source-to-surface distance, field size, temperature, and orientation CFs for cobalt-60 and 9 MV beams. Most of the simulation results were within 4% of the measurements. The results suggest against the use of the EDD5 in axial angles beyond ± 50° and exceeding the range 0° to +50° tilt angle at 9 MV. Outside these ranges, although the EDD5 can be used for accurate in vivo dosimetry at 9 MV, its CF variations were found to be 1.5-7.1 times larger than the EDP10 and, therefore, should be applied carefully. Finally, the MCNP diode models are sufficiently reliable tools for independent verification of potentially inaccurate measurements.